US10746830B2 - Systems and methods for hybrid slice encoding in three-dimensional magnetic resonance imaging - Google Patents

Abstract

Methods and systems are provided for hybrid slice encoding. In one embodiment, a method for magnetic resonance imaging comprises, during a scan with a pulse sequence, sampling k-space linearly for a predetermined number of echoes, and sampling k-space centrically for remaining echoes of the pulse sequence. In this way, blurriness along the slice direction may be reduced for 3D fast spin echo imaging.

Description

FIELD

Embodiments of the subject matter disclosed herein relate to medical imaging, such as magnetic resonance imaging (MRI), and more particularly, to slice encoding for MRI.

BACKGROUND

Magnetic resonance imaging (MRI) is a medical imaging modality that can create pictures of the inside of a human body without using x-rays or other ionizing radiation. MRI uses a superconducting magnet to create a strong, uniform, static magnetic field. When a human body, or part of a human body, is placed in the magnetic field, the nuclear spins associated with the hydrogen nuclei in tissue water become polarized, wherein the magnetic moments associated with these spins become preferentially aligned along the direction of the magnetic field, resulting in a small net tissue magnetization along that axis. MRI systems also include gradient coils that produce smaller-amplitude, spatially-varying magnetic fields with orthogonal axes to spatially encode the MR signal by creating a signature resonance frequency at each location in the body. Radio frequency (RF) coils are then used to create pulses of RF energy at or near the resonance frequency of the hydrogen nuclei, which add energy to the nuclear spin system. As the nuclear spins relax back to their rest energy state, they release the absorbed energy in the form of an RF signal. This signal is detected by the MRI system and is transformed into an image using reconstruction algorithms.

BRIEF DESCRIPTION

In one embodiment, a method for magnetic resonance imaging comprises, during a scan with a pulse sequence, sampling k-space linearly for a predetermined number of echoes, and sampling k-space centrically for remaining echoes of the pulse sequence. In this way, blurring along the slice direction may be reduced for three-dimensional fast spin echo imaging.

It should be understood that the brief description above is provided to introduce in simplified form a selection of concepts that are further described in the detailed description. It is not meant to identify key or essential features of the claimed subject matter, the scope of which is defined uniquely by the claims that follow the detailed description. Furthermore, the claimed subject matter is not limited to implementations that solve any disadvantages noted above or in any part of this disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS

The present invention will be better understood from reading the following description of non-limiting embodiments, with reference to the attached drawings, wherein below:

FIG. 1 is a block diagram of an MRI system according to an exemplary embodiment;

FIGS. 2 and 3 show diagrams illustrating an example pulse sequence for three-dimensional arterial spin labeling (3D ASL) according to an embodiment;

FIG. 4 shows a set of graphs and images illustrating control and labeling pulse sequences for generating a perfusion weighted image according to an exemplary embodiment;

FIG. 5 shows a graph illustrating phantom magnitude data acquired without slice encoding;

FIG. 6 shows a graph illustrating phantom phase data (in degrees) acquired without slice encoding;

FIG. 7 shows a graph illustrating an example centric slice encoding;

FIG. 8 shows a high-level flow chart illustrating an example method for a three-dimensional scan with hybrid slice encoding according to an embodiment;

FIG. 9 shows a high-level flow chart illustrating an example method for hybrid slice encoding according to an embodiment;

FIG. 10 shows a graph illustrating hybrid slice encoding according to an exemplary embodiment;

FIG. 11 shows a graph illustrating an example centric slice encoding with echo skipping;

FIG. 12 shows a graph illustrating an example linear partial slice encoding;

FIG. 13 shows a set of images acquired with different slice encoding techniques;

FIG. 14 shows a set of proton density images acquired with centric slice encoding and with hybrid slice encoding; and

FIG. 15 shows a set of perfusion weighted images acquired with centric slice encoding and with hybrid slice encoding.

DETAILED DESCRIPTION

The following description relates to various embodiments of reducing image blurring for medical imaging systems. In particular, systems and methods are provided for reducing blurring along a slice direction for images acquired with a medical imaging system, such as the MRI system depicted inFIG. 1. In particular, blurring along the slice direction occurs for images acquired using three-dimensional arterial spin labeling (3D ASL) or other fast spin echo (FSE) techniques, as depicted inFIGS. 2-4. The blurring occurs due to sampling the center of k-space, when the signal magnitude is strongest as depicted inFIG. 5, while the signal phase varies substantially, as depicted inFIG. 6. In particular, FSE imaging techniques such as 3D ASL use centric slice encoding, illustrated inFIG. 7, which exacerbates the blurring along the slice direction. A method for 3D MRI, such as the method depicted inFIG. 8, includes using a hybrid slice encoding technique. A method for a hybrid slice encoding technique, such as the method depicted inFIG. 9, includes using a linear slice encoding during the initial phase variation before switching to a centric slice encoding, as depicted inFIG. 10. In this way, the large phase variation at the k-space center is avoided, thus reducing slice-direction blurring, while the signal from the beginning echoes is kept, thus preserving the signal-to-noise ratio (SNR). Other slice encoding techniques for avoiding the phase variation at the k-space center include centric slice encoding with echo skipping, as depicted inFIG. 11, and linear partial slice encoding, as depicted inFIG. 12. Example images acquired with the slice encoding techniques described herein, such as the images depicted inFIGS. 13-15, demonstrate the efficacy of the hybrid slice encoding in comparison with conventional centric slice encoding, centric slice encoding with echo skipping, and linear partial slice encoding.

FIG. 1 illustrates a magnetic resonance imaging (MRI)apparatus10 that includes a magnetostaticfield magnet unit12, agradient coil unit13, anRF coil unit14, an RFbody coil unit15, a transmit/receive (T/R)switch20, anRF driver unit22, a gradientcoil driver unit23, adata acquisition unit24, acontroller unit25, apatient bed26, a data processing unit31, anoperating console unit32, and adisplay unit33. TheMRI apparatus10 transmits electromagnetic pulse signals to asubject16 placed in animaging space18 with a magnetostatic field formed to perform a scan for obtaining magnetic resonance (MR) signals from thesubject16 to reconstruct an image of a slice of thesubject16 based on the MR signals thus obtained by the scan.

The magnetostaticfield magnet unit12 includes, for example, typically an annular superconducting magnet, which is mounted within a toroidal vacuum vessel. The magnet defines a cylindrical space surrounding thesubject16, and generates a constant primary magnetostatic field B₀.

TheMRI apparatus10 also includes thegradient coil unit13 that generates a gradient magnetic field in theimaging space18 so as to provide the MR signals received by theRF coil unit14 with three-dimensional positional information. Thegradient coil unit13 includes three gradient coil systems, each of which generates a gradient magnetic field which includes into one of three spatial axes perpendicular to each other, and generates a gradient field in each frequency-encoding direction, phase-encoding direction, and slice-selection direction in accordance with the imaging condition. More specifically, thegradient coil unit13 applies a gradient field in the slice-selection direction (or scan direction) of thesubject16, to select the slice; and theRF coil unit14 transmits an RF pulse to a selected slice of thesubject16 and excites it. Thegradient coil unit13 also applies a gradient field in the phase-encoding direction of thesubject16 to phase encode the MR signals from the slice excited by the RF pulse. Thegradient coil unit13 then applies a gradient field in the frequency-encoding direction of thesubject16 to frequency encode the MR signals from the slice excited by the RF pulse.

TheRF coil unit14 is disposed, for example, to enclose the region to be imaged of thesubject16. In the static magnetic field space orimaging space18 where a static magnetic field B₀is formed by the magnetostaticfield magnet unit12, theRF coil unit14 transmits, based on a control signal from thecontroller unit25, an RF pulse that is an electromagnetic wave to thesubject16 and thereby generates a high-frequency magnetic field B₁. This excites a spin of protons in the slice to be imaged of thesubject16. TheRF coil unit14 receives, as an MR signal, the electromagnetic wave generated when the proton spin thus excited in the slice to be imaged of thesubject16 returns into alignment with the initial magnetization vector. In one embodiment, theRF coil unit14 may transmit and receive an RF pulse using the same RF coil. In another embodiment, theRF coil unit14 may be used for only receiving the MR signals, but not transmitting the RF pulse.

The RFbody coil unit15 is disposed, for example, to enclose theimaging space18, and produces RF magnetic field pulses B₁orthogonal to the main magnetic field B₀produced by the magnetostaticfield magnet unit12 within theimaging space18 to excite the nuclei. In contrast to theRF coil unit14, which may be easily disconnected from theMR apparatus10 and replaced with another RF coil unit, the RFbody coil unit15 is fixedly attached and connected to theMRI apparatus10. Furthermore, whereas local coils such as those comprising theRF coil unit14 can transmit to or receive signals from only a localized region of thesubject16, the RFbody coil unit15 generally has a larger coverage area and can be used to transmit or receive signals to the whole body of thesubject16. Using receive-only local coils and transmit body coils provides a uniform RF excitation and good image uniformity at the expense of high RF power deposited in thesubject16. For a transmit-receive local coil, the local coil provides the RF excitation to the region of interest and receives the MR signal, thereby decreasing the RF power deposited in thesubject16. It should be appreciated that the particular use of theRF coil unit14 and/or the RFbody coil unit15 depends on the imaging application.

The T/R switch20 can selectively electrically connect the RFbody coil unit15 to thedata acquisition unit24 when operating in a receive mode, and to theRF driver unit22 when operating in transmit mode. Similarly, the T/R switch20 can selectively electrically connect theRF coil unit14 to thedata acquisition unit24 when theRF coil unit14 operates in receive mode, and to theRF driver unit22 when operating in transmit mode. When theRF coil unit14 and the RFbody coil unit15 are both used in a single scan, for example if theRF coil unit14 is configured to receive MR signals and the RFbody coil unit15 is configured to transmit RF signals, then the T/R switch20 may direct control signals from theRF driver unit22 to the RFbody coil unit15 while directing received MR signals from theRF coil unit14 to thedata acquisition unit24. The coils of the RFbody coil unit15 may be configured to operate in a transmit-only mode or a transmit-receive mode. The coils of the localRF coil unit14 may be configured to operate in a transmit-receive mode or a receive-only mode.

TheRF driver unit22 includes a gate modulator (not shown), an RF power amplifier (not shown), and an RF oscillator (not shown) that are used to drive theRF coil unit14 or the RFbody coil unit15 and form a high-frequency magnetic field in theimaging space18. TheRF driver unit22 modulates, based on a control signal from thecontroller unit25 and using the gate modulator, the RF signal received from the RF oscillator into a signal of predetermined timing having a predetermined envelope. The RF signal modulated by the gate modulator is amplified by the RF power amplifier and then output to theRF coil unit14 or the RFbody coil unit15.

The gradientcoil driver unit23 drives thegradient coil unit13 based on a control signal from thecontroller unit25 and thereby generates a gradient magnetic field in theimaging space18. The gradientcoil driver unit23 includes three systems of driver circuits (not shown) corresponding to the three gradient coil systems included in thegradient coil unit13.

Thedata acquisition unit24 includes a preamplifier (not shown), a phase detector (not shown), and an analog/digital converter (not shown) used to acquire the MR signals received by theRF coil unit14. In thedata acquisition unit24, the phase detector phase detects, using the output from the RF oscillator of theRF driver unit22 as a reference signal, the MR signals received from theRF coil unit14 and amplified by the preamplifier, and outputs the phase-detected analog MR signals to the analog/digital converter for conversion into digital signals. The digital signals thus obtained are output to the data processing unit31.

TheMRI apparatus10 includes a table26 for placing the subject16 thereon. The subject16 may be moved inside and outside theimaging space18 by moving the table26 based on control signals from thecontroller unit25.

Thecontroller unit25 includes a computer and a recording medium on which a program to be executed by the computer is recorded. The program when executed by the computer causes various parts of the apparatus to carry out operations corresponding to pre-determined scanning. The recording medium may comprise, for example, a ROM, flexible disk, hard disk, optical disk, magneto-optical disk, CD-ROM, or non-volatile memory card. Thecontroller unit25 is connected to theoperating console unit32 and processes the operation signals input to theoperating console unit32 and furthermore controls the table26,RF driver unit22, gradientcoil driver unit23, anddata acquisition unit24 by outputting control signals to them. Thecontroller unit25 also controls, to obtain a desired image, the data processing unit31 and thedisplay unit33 based on operation signals received from the operatingconsole unit32.

The operatingconsole unit32 includes user input devices such as, as non-limiting examples, a keyboard and a mouse. The operatingconsole unit32 is used by an operator, for example, to input such data as an imaging protocol and to set a region where an imaging sequence is to be executed. The data about the imaging protocol and the imaging sequence execution region are output to thecontroller unit25.

The data processing unit31 includes a computer and a recording medium on which a program to be executed by the computer to perform pre-determined data processing is recorded. The data processing unit31 is connected to thecontroller unit25 and performs data processing based on control signals received from thecontroller unit25. The data processing unit31 is also connected to thedata acquisition unit24 and generates spectrum data by applying various imaging processing operations to the MR signals output from thedata acquisition unit24.

Thedisplay unit33 includes a display device and displays an image on the display screen of the display device based on control signals received from thecontroller unit25. Thedisplay unit33 displays, for example, an image regarding an input item about which the operator inputs operation data from the operatingconsole unit32. Thedisplay unit33 also displays a slice image of the subject16 generated by the data processing unit31.

FIG. 2 shows a diagram illustrating anexample pulse sequence200 for three-dimensional arterial spin labeling according to an embodiment. Thepulse sequence200 includes a spin preparation sequence or more simply apreparation sequence202, followed by a three-dimensional spiral fast spin echo (FSE)sequence204, also referred to as a readout sequence. During thepreparation sequence202, theRF pulse sequence210 includespre-saturation pulses211, aselective inversion pulse212, alabeling pulse213, as well as saturation andbackground suppression pulses214, while theGz pulse sequence220 includes a pulse corresponding to thelabeling pulse213.

During the 3Dspiral FSE sequence204 or thereadout sequence204, theRF pulse sequence210 includes a first 90°RF pulse215 followed by a series of refocusingRF pulses216, while theGz pulse sequence220 includes a plurality of slice encoding pulses. TheGy pulse sequence230 and theGz pulse sequence240 both include a plurality of echoes, wherein each echo follows a corresponding refocusing RF pulse of theRF pulse sequence210, as depicted.

FIG. 3 shows a higher-level diagram illustrating anexample pulse sequence300 for three-dimensional arterial spin labeling. In particular, thepulse sequence300 includes a labeling andbackground suppression sequence305, corresponding to thepreparation sequence202 described hereinabove, followed by a 3DFSE echo train310, corresponding to the 3Dspiral FSE sequence204 described hereinabove. The 3DFSE echo train310 is followed by a second preparation sequence comprising a control andbackground suppression sequence315, which in turn is followed by another 3DFSE echo train320. Theentire pulse sequence300 may be repeated for a number of excitations and interleaves. As known in the art of ASL, the image from dataset acquired during the 3D FSEecho train sequence310 comprises a labeling image while the image from dataset acquired during the 3D FSEecho train sequence320 comprises a control image, and a perfusion weighted image is obtained by subtracting the control image and the labeling image.

For example,FIG. 4 shows a set of graphs andimages400 illustrating acontrol pulse sequence410 and alabeling pulse sequence420 for generating a perfusionweighted image430. The average amplitude of theRF pulse sequence411 of thecontrol pulse sequence410 is zero, while the average gradient of theGz pulse sequence412 of thecontrol pulse sequence410 is also zero over time. Anexample control image415 is reconstructed from k-space data acquired during thecontrol sequence410. In contrast, the average amplitude of theRF pulse sequence421 of thelabeling pulse sequence420 is non-zero, while the average gradient of theGz pulse sequence422 of the labeling pulse sequence is also non-zero over time. Anexample labeling image425 is reconstructed from k-space data acquired during thelabeling sequence420. The perfusion weighted (PW)image430 is generated by subtracting thelabeling image425 from thecontrol image415.

Blurring along the slice direction is observable in images acquired via 3D ASL imaging as described hereinabove, and in 3D FSE imaging in general.

FIG. 5 shows agraph500 illustrating the magnitude data of MR signal acquired from a phantom without slice encoding. Thegraph500 includes aplot505 of magnitude data corresponding to control sequences (in solid line) and aplot510 of magnitude data corresponding to labeling sequences (in dashed line) over a plurality of echoes. For both the control and the labeling sequences, the signal decays substantially over the readout echo train. Meanwhile,FIG. 6 shows acorresponding graph600 illustrating the phase data of the MR signal acquired from the phantom without slice encoding, including aplot605 of the phase data corresponding to control sequences (in solid line) and aplot610 of phase data corresponding to labeling sequences (in dashed line) over the plurality of echoes. As observed ingraph600, the phase illustrated by bothplots605 and610 jumps substantially within the first few echoes.

This clearly observable phase jump, which occurs because the echo train has not reached a steady state yet, is responsible for the blurring along the slice direction. More specifically, the phase jump in conjunction with a centric slice encoding is responsible for the blurring along the slice direction.

As an example,FIG. 7 shows agraph700 illustrating centric slice encoding. Thegraph700 includes an illustration of thesignal decay702 during the readout sequence. Thegraph700 further depicts a plurality ofpoints708 wherein eachpoint707 indicates the slice encoding for a given echo. The slice encoding number corresponds to the locations of the k-space sampled with the given echo. If the k-space center is represented by slice encoding number “0” as shown, then the lower the absolute value of the slice encoding number, the region closer to the center of k-space center is sampled. For the centric slice encoding scheme, k-space is sampled beginning at the center of k-space (i.e., at 0 slice encoding) at the 0^thecho709, then moves to positive k-space for the 1^stecho710, negative k-space for the 2^ndecho711, and so on in an interleaved fashion until the k-space is filled.

As mentioned above, sampling the center of k-space as illustrated byFIG. 7 during the phase jump depicted inFIG. 6 causes blurring along the slice direction. As described further herein, the blurring along the slice direction may be reduced or eliminated by utilizing a hybrid slice encoding scheme comprising a combination of linear slice encoding and centric slice encoding.

FIG. 8 shows a high-level flow chart illustrating anexample method800 for a three-dimensional scan with hybrid slice encoding according to an embodiment.Method800 is described with regard to the systems and components ofFIG. 1, though it should be appreciated that themethod800 may be implemented with other systems and components without departing from the scope of the present disclosure.Method800 may be stored as executable instructions in non-transitory memory of a computing device of theMRI apparatus10, for example.

Method800 begins at805. At805,method800 begins a 3D scan. The 3D scan may comprise a 3D ASL scan, as a non-limiting example, or another type of 3D scan that includes a 3D FSE pulse sequence as described hereinabove with regard toFIG. 2.

In the case that the 3D scan comprises a 3D ASL scan, at810,method800 optionally controls the RF coil and the gradient coils with preparation waveforms for a control or a labeling pulse sequence. For example,method800 may control the RF coil and one or more gradient coils to generate pre-saturation pulses, selective inversion pulses, labeling or control pulses, saturation pulses, and background suppression pulses as described hereinabove with regard toFIGS. 2-4.

After optionally performing the preparation sequence at810,method800 continues to815. At815,method800 controls the RF coil and the gradient coils with a FSE pulse sequence, such as the FSE pulse sequence described hereinabove with regard toFIG. 2, with a hybrid slice encoding. The hybrid slice encoding comprises linear slice encoding for a predetermined number of echoes in the FSE pulse sequence, followed by centric slice encoding for the remaining echoes of the FSE pulse sequence. An example method for hybrid slice encoding is described further herein with regard toFIG. 9.

At820,method800 determines whether the scan is complete. If the scan is not complete (“NO”),method800 returns to810, whereinmethod800 controls the coils with preparation waveforms for control or labeling. In particular,method800 uses control or labeling waveforms according to which pulse sequence was utilized in the previous iteration, such that the scan alternates between control and labeling sequences as described hereinabove with regard toFIGS. 2-4. For example, during a first iteration,method800 controls the coils with a labeling pulse sequence, and during a second iteration following the first iteration,method800 controls the coils with a control pulse sequence, and so on untilmethod800 determines that the scan is complete at820.

Referring again to820, once the scan is complete (“YES”),method800 continues to825, whereinmethod800 ends the 3D scan. Continuing at830,method800 reconstructs one or more images from the scan data acquired during the 3D scan. Reconstructing one or more images may comprise reconstructing control and labeling images and generating a PW image from the difference, as described hereinabove, or reconstructing another image from the scan data. At835,method800 outputs the one or more images reconstructed at830, for example for display via adisplay unit33 or to a storage device for subsequent retrieval. An image acquired with hybrid slice encoding as described herein exhibits less blurring along the slice direction, especially when compared to an image acquired with centric slice encoding or other slice encoding techniques.Method800 then returns.

FIG. 9 shows a high-level flow chart illustrating anexample method900 for hybrid slice encoding according to an embodiment.Method900 is described with reference to the systems and components ofFIG. 1, though it should be appreciated that themethod900 may be implemented with other systems and components without departing from the scope of the present disclosure.

Method900 begins at905. At905,method900 begins a 3D FSE sequence. That is,method900 begins controlling one or more RF coils and gradient coils according to a 3D FSE pulse sequence, for example as described hereinabove with regard toFIG. 2.

During the 3D FSE sequence, at910,method900 controls the slice encoding gradient with linear slice encoding for a predetermined number of echoes. The number of echoes may be predetermined such that the slice encoding is linear while a large phase variation exists at the beginning of the echo train. As an illustrative example,FIG. 10 shows agraph1000 illustrating a hybrid slice encoding, including alinear slice encoding1003 at the beginning of the echo train.

In some examples, the linear slice encoding may range from a negative value to a positive value sequentially in k-space. For example, for k-space ranging from −kmax to +kmax, the slice encoding may begin at a negative value −kb and move sequentially to a positive value +kb (i.e., from −kb, to −kb+1, −kb+2, . . . to +kb−1, +kb), wherein kb<kmax. Furthermore, the range [−kb, +kb] of the linear slice encoding may be selected or predetermined such that the linear slice encoding crosses zero, or the center of k-space, when the large phase variation or phase jump at the beginning of the echo train ends. For example, as depicted inFIG. 6, the large phase variation settles down around the fourth echo (echo index three). As depicted inFIG. 10, thelinear slice encoding1003 ranges from slice encoding number −4 sequentially to 4 (e.g., from −4 to −3, −2, −1, 0, 1, 2, 3, 4), with the linear slice encoding cross theslice encoding number 0 at the fourth echo (echo index three). Thus, in the example depicted inFIG. 10, the range [−kb,+kb] is set to encoding number [−4,+4] such that thelinear slice encoding1003 uses the first eight echoes of the echo train. The number of echoes at which the linear slice encoding ends may be considered a predetermined number of echoes.

Alternatively, in one or more examples, sampling k-space linearly (or sequentially) for the predetermined number of echoes comprises encoding a positive slice encoding number for a first echo of the pulse sequence, and linearly decreasing a slice encoding number for encoding subsequent echoes of the pulse sequence until the predetermined number of echoes is reached.

Continuing at915,method900 controls the slice encoding gradient with centric slice encoding for the remaining echoes. That is, for the remaining echoes of the echo train in the 3D FSE sequence, the slice encoding is centric for the slice ranges [−kmax, −kb) and (+kb, +kmax], e.g., from −kb−1, to +kb+1, to −kb−2, to +kb+2, etc. In the example depicted inFIG. 10, thecentric slice encoding1004 begins at slice encoding number −5 after thelinear slice encoding1003 ends, then goes to sliceencoding number +5, encoding number −6, encoding number +6, and so on. After covering the remainder of k-space for the remainder of the echo train with centric slice encoding,method900 continues to920. At920,method900 ends the 3D FSE sequence.Method900 then returns.

Thus, a method for reducing blurring along the slice direction for 3D MRI includes sampling k-space with a linear (or sequential) slice encoding scheme for a predetermined number of echoes, followed by a centric slice encoding scheme for the remaining echoes of the echo train.

Other slice encoding techniques may be used to address the blurring along the slice direction. For example,FIG. 11 shows agraph1100 illustratingcentric slice encoding1108 with echo skipping. In particular, thecentric slice encoding1108 begins at k-space center after skipping a predetermined number of echoes, creating aslice encoding gap1102 in the beginning of the echo train. Skipping the predetermined number of echoes helps reduce the blurring, but may reduce the signal-to-noise ratio (SNR), or more specifically causes signal reduction per the T2 decay.

As another example of a slice encoding technique to reduce blurring along the slice direction,FIG. 12 shows agraph1200 illustrating linearpartial slice encoding1208. As depicted, thelinear slice encoding1208 begins at −kb, similar to thelinear slice encoding1003 of the hybrid slice encoding technique, but extends linearly (or sequentially) to +kmax. As depicted, no echoes are slice encoded in the range [−kmax, −kb), hence the k-space coverage is asymmetric or partial. The negative portions of k-space that are not filled may be synthesized or interpolated from the acquired data during image reconstruction, for example by using partial Fourier imaging techniques.

FIG. 13 shows a set ofimages1300 acquired with different slice encoding techniques.Image1305 is acquired with centric slice encoding, and thus exhibits significant blurring along the slice direction.Image1310 is acquired with hybrid slice encoding as described herein, wherein the linear segment of the hybrid slice encoding lasts for sixteen echoes.Image1315 is acquired with hybrid slice encoding, wherein the linear segment of the hybrid slice encoding lasts for eight echoes. The blurriness along the slice direction inimages1310 and1315 is reduced with respect toimage1305, though the SNR of theimage1310 may be undesirably low in comparison toimage1315.Image1320 is acquired with centric slice encoding, wherein the first four echoes are skipped.Image1325 is acquired with centric slice encoding, wherein the first six echoes are skipped. While the echo skipping for bothimages1320 and1325 reduces the blurring along the slice direction by avoiding the large phase variation at the beginning of the FSE pulse sequence, the strong signal from the early echoes is missing from the dataset and thus the SNR is substantially lower with respect to theimages1310 and1315. Thus, of theimages1300, theimage1315 is the most balanced with regard to blurring and SNR.

As another example,FIG. 14 shows a set ofproton density images1400 acquired with centric slice encoding versus with hybrid slice encoding. In particular,image1405 shows a coronal view acquired with centric slice encoding, whileimage1407 shows a coronal view of the same subject acquired with hybrid slice encoding.Image1410 shows a sagittal view acquired with centric slice encoding, whileimage1412 shows a sagittal view acquired with hybrid slice encoding. Theimages1407 and1412 exhibit a substantial reduction in blurring along the slice direction in comparison with the correspondingimages1405 and1410.

As yet another example,FIG. 15 shows a set of perfusion weightedimages1500 acquired with centric slice encoding versus with hybrid slice encoding. In particular,image1505 shows a coronal view acquired with centric slice encoding, whereasimage1507 shows a coronal view acquired with hybrid slice encoding.Image1510 shows a sagittal view acquired with centric slice encoding, whileimage1512 shows a sagittal view acquired with hybrid slice encoding. The reduction in blurring along the slice direction is even more substantial in theimages1507 and1512 in comparison to theproton density images1407 and1412, indicating that the hybrid slice encoding technique described herein is especially advantageous for perfusion weighted imaging.

A technical effect of the disclosure includes encoding slices linearly and centrically during magnetic resonance imaging. Another technical effect of the disclosure includes the reconstruction and display of images with reduced blurring in the slice direction.

Thus, provided herein are methods and systems for magnetic resonance (MR) imaging. In a first example method, during a scan with a pulse sequence, k-space may be sampled linearly for a predetermined number of echoes, and k-space may be sampled centrically for remaining echoes of the pulse sequence. In a second example method which may optionally include the first example method, sampling the k-space linearly for the predetermined number of echoes may comprise acquiring MR data along a first slice encoding which is offset from a k-space center for a first echo of the pulse sequence, and then acquiring MR data sequentially for slice encodings until the predetermined number of echoes is reached. In one or more example methods, sampling the k-space linearly for the predetermined number of echoes may further comprise acquiring MR data along a slice encoding at the k-space center. In another example method which may include one or more of the features described above, phase variation in the MR data acquired along the slice encoding at the k-space center may be less than phase variation in the MR data acquired along the first slice encoding. Furthermore, in one or more example methods, sampling the k-space centrically for the remaining echoes of the pulse sequence may comprise acquiring MR data alternating between slice encodings at both sides of the slice encoding at the k-space center for each echo of the remaining echoes. Optionally, one or more of the example methods may include a pulse sequence that comprises a three-dimensional fast spin echo pulse sequence, and controlling a radiofrequency coil and a plurality of gradient coils according to the pulse sequence multiple times during the scan, wherein the k-space is sampled linearly and centrically each time.

In a further example method, which may optionally include one or more of the method features described above, a method for magnetic resonance imaging may comprise acquiring k-space data with a linear k-space trajectory and a centric k-space trajectory, and reconstructing an image from the acquired k-space data. In another example method which optionally includes one or more of the method features described above, acquiring the k-space data with the linear k-space trajectory may comprise acquiring the k-space data sequentially starting from a slice encoding which is offset from a k-space center.

Acquiring the k-space data with the centric k-space trajectory may comprise acquiring the k-space data alternating between slice encodings at both sides of a slice encoding at the k-space center, in one or more example methods. Moreover, in at least one example method, the linear k-space trajectory may cross the center of the k-space after a phase jump of a pulse sequence is substantially reduced. Further, in one or more examples, acquiring the k-space data with the linear k-space trajectory may be for a predetermined number of echoes. Additionally, in one or more examples, the centric k-space trajectory may be switched to after the predetermined number of echoes.

In at least one example method, which may optionally include one or more features of the above-described methods, comprises acquiring additional k-space data with the linear k-space trajectory and the centric k-space trajectory, reconstructing a second image from the additional k-space data, and generating a perfusion weighted image from a difference between the image and the second image.

In an example MR imaging system, the MR imaging system may comprise, a radiofrequency (RF) coil, a gradient coil unit including a plurality of gradient coils, and a processor communicatively coupled to the RF coil and the gradient coil unit and configured with instructions that when executed cause the processor to carry out any one of the example methods described above. For example, the instructions may cause the processor to, during a scan with a pulse sequence, control the RF coil and the gradient coils to sample k-space linearly for a predetermined number of echoes of the pulse sequence and sample the k-space centrically for remaining echoes of the pulse sequence. In one or more examples, sampling the k-space linearly for the predetermined number of echoes may comprise, acquiring MR data along a first slice encoding which is offset from a k-space center for a first echo of the pulse sequence, and then acquiring MR data sequentially for slice encodings until the predetermined number of echoes is reached. Further, in at least one example, sampling the k-space linearly for the predetermined number of echoes may further comprise acquiring MR data along a slice encoding at the k-space center. In one or more example systems, phase variation in the MR data acquired along the slice encoding at the k-space center may be less than phase variation in the MR data acquired along the first slice encoding. Further, in one or more examples, the processor may be further configured to reconstruct an image from data acquired during the scan. In at least one example system, sampling the k-space centrically for the remaining echoes of the pulse sequence may comprise acquiring MR data alternating between slice encodings at both sides of the slice encoding at the k-space center for each echo of the remaining echoes. Further, in at least one example, phase variation in MR data acquired along a slice encoding may be substantially reduced after the predetermined number of echoes.

As used herein, an element or step recited in the singular and proceeded with the word “a” or “an” should be understood as not excluding plural of said elements or steps, unless such exclusion is explicitly stated. Furthermore, references to “one embodiment” of the present invention are not intended to be interpreted as excluding the existence of additional embodiments that also incorporate the recited features. Moreover, unless explicitly stated to the contrary, embodiments “comprising,” “including,” or “having” an element or a plurality of elements having a particular property may include additional such elements not having that property. The terms “including” and “in which” are used as the plain-language equivalents of the respective terms “comprising” and “wherein.” Moreover, the terms “first,” “second,” and “third,” etc. are used merely as labels, and are not intended to impose numerical requirements or a particular positional order on their objects.

This written description uses examples to disclose the invention, including the best mode, and also to enable a person of ordinary skill in the relevant art to practice the invention, including making and using any devices or systems and performing any incorporated methods. The patentable scope of the invention is defined by the claims, and may include other examples that occur to those of ordinary skill in the art. Such other examples are intended to be within the scope of the claims if they have structural elements that do not differ from the literal language of the claims, or if they include equivalent structural elements with insubstantial differences from the literal languages of the claims.

Claims (18)

The invention claimed is:

1. A method for magnetic resonance (MR) imaging, comprising:

during a scan with a pulse sequence, sampling k-space linearly for a predetermined number of echoes, wherein sampling the k-space linearly for the predetermined number of echoes comprises acquiring MR data along a first slice encoding which is offset from a k-space center for a first echo of the pulse sequence, and then acquiring MR data sequentially for slice encodings until the predetermined number of echoes is reached; and

sampling k-space centrically for remaining echoes of the pulse sequence.

2. The method ofclaim 1, wherein sampling the k-space linearly for the predetermined number of echoes further comprises acquiring MR data along a slice encoding at the k-space center.

3. The method ofclaim 2, wherein phase variation in the MR data acquired along the slice encoding at the k-space center is less than phase variation in the MR data acquired along the first slice encoding.

4. The method ofclaim 1, wherein sampling the k-space centrically for the remaining echoes of the pulse sequence comprises acquiring MR data alternating between slice encodings at both sides of a slice encoding at the k-space center for each echo of the remaining echoes.

5. The method ofclaim 1, wherein the pulse sequence comprises a three-dimensional fast spin echo pulse sequence, and further comprising controlling a radiofrequency coil and a plurality of gradient coils according to the pulse sequence multiple times during the scan, wherein the k-space is sampled linearly and centrically each time.

6. A method for magnetic resonance imaging, comprising:

acquiring k-space data with a linear k-space trajectory for a predetermined number of echoes in a pulse sequence, the predetermined number of echoes selected such that the linear k-space trajectory crosses a k-space center when a phase jump of the pulse sequence settles, and a centric k-space trajectory for remaining echoes of the pulse sequence; and

reconstructing an image from the acquired k-space data.

7. The method ofclaim 6, wherein acquiring the k-space data with the linear k-space trajectory comprises acquiring the k-space data sequentially starting from a slice encoding which is offset from the k-space center.

8. The method ofclaim 7, wherein acquiring the k-space data with the centric k-space trajectory comprises acquiring the k-space data alternating between slice encodings at both sides of a slice encoding at the k-space center.

9. The method ofclaim 7, wherein the linear k-space trajectory crosses the k-space center when the phase jump of the pulse sequence is settled.

10. The method ofclaim 7, wherein the linear k-space trajectory ranges from a negative k-space value to a positive k-space value.

11. The method ofclaim 10, wherein an absolute value of the negative k-space value is equal to the positive k-space value, and wherein the linear k-space trajectory begins at the negative k-space value or the positive k-space value.

12. The method ofclaim 6, further comprising acquiring additional k-space data with the linear k-space trajectory and the centric k-space trajectory, reconstructing a second image from the additional k-space data, and generating a perfusion weighted image from a difference between the image and the second image.

13. An MR imaging system, comprising:

a radiofrequency (RF) coil;

a gradient coil unit including a plurality of gradient coils;

a processor communicatively coupled to the RF coil and the gradient coil unit and configured with instructions stored in non-transitory memory that when executed cause the processor to:

during a scan with a pulse sequence, control the RF coil and the plurality of gradient coils to sample k-space linearly for a predetermined number of echoes of the pulse sequence and sample the k-space centrically for remaining echoes of the pulse sequence, wherein sampling the k-space linearly for the predetermined number of echoes comprises acquiring MR data along a first slice encoding which is offset from a k-space center for a first echo of the pulse sequence, and then acquiring MR data sequentially for slice encodings until the predetermined number of echoes is reached.

14. The MR imaging system ofclaim 13, wherein sampling the k-space linearly for the predetermined number of echoes further comprises acquiring MR data along a slice encoding at the k-space center.

15. The MR imaging system ofclaim 14, wherein phase variation in the MR data acquired along the slice encoding at the k-space center is less than phase variation in the MR data acquired along the first slice encoding.

16. The MR imaging system ofclaim 13, wherein the processor is further configured to reconstruct an image from data acquired during the scan.

17. The MR imaging system ofclaim 13, wherein sampling the k-space centrically for the remaining echoes of the pulse sequence comprises acquiring MR data alternating between slice encodings at both sides of a slice encoding at the k-space center for each echo of the remaining echoes.

18. The MR imaging system ofclaim 13, wherein phase variation in MR data acquired along a slice encoding is substantially reduced after the predetermined number of echoes.

Images